Flogging a dead norm? - Center for Science and Technology Policy ...

Area (2007) 39.2, 000– 000

Flogging a dead norm? Newspaper coverage of anthropogenic climate change in the United States and United Kingdom from 2003 to 2006 Blackwell Publishing Ltd

Maxwell T Boykoff Environmental Change Institute, Oxford University Centre for the Environment, Oxford OX1 3QY Email: [email protected] Revised manuscript received 7 June 2007 The journalistic norm of ‘balanced’ reporting (giving roughly equal coverage to both sides in any significant dispute) is recognised as both useful and problematic in communicating emerging scientific consensus on human attribution for global climate change. Analysis of the practice of this norm in United States (US) and United Kingdom (UK) newspaper coverage of climate science between 2003 and 2006 shows a significant divergence from scientific consensus in the US in 2003 – 4, followed by a decline in 2005– 6, but no major divergence in UK reporting. These findings inform ongoing considerations about the spatially-differentiated media terms and conditions through which current and future climate policy is negotiated and implemented. Key words: United States, United Kingdom, climate science, mass media, policy, content analysis

Introduction The professionalised and institutionalised journalistic norm of ‘balanced reporting’ is generally considered to be a vital tool in carrying out ‘objective’ reporting that provides ‘both sides in any significant dispute with roughly equal attention’ (Entman 1989, 30). This norm guides how many news stories are framed and covered (Cunningham 2003) and can provide a valuable ‘fairness check’ for reporters who have neither the time nor the scientific understanding to verify the legitimacy of competing claims about any given issue (Gamson and Modigliani 1989; Dunwoody and Peters 1992). While effective in many cases, the employment of this norm to issues such as anthropogenic climate change can be problematic (Boykoff and Boykoff 2004). Rather than providing accurate information, ‘balanced’ reporting may instead perpetrate informational bias regarding scientific opinions on human contributions to climate change. This paper seeks to assess the potential for such bias by exploring

the extent to which ‘balanced’ media coverage (commonly called ‘he said/she said’ reporting) of anthropogenic climate change remains a significant feature in United States (US) and United Kingdom (UK) reporting of this issue. Scientific understanding of the causes of climate change has evolved markedly in recent decades. Particularly in the last dozen years, reports and findings have signalled a broad scientific consensus that human actions are contributing to modern climate change – despite lingering uncertainties regarding the extent of attribution. For instance, the recent United Nations-sponsored Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) from Working Group I (WGI) states that ‘Most of the observed increase in globally averaged temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations’ (IPCC 2007, 8; emphasis added). Fielding over 30 000 comments from experts and governments, this multi-stage peerreview and consensus-building process represents a

Area Vol. 39 No. 2, pp. 000 –000, 2007 ISSN 0004-0894 © The Author. Journal compilation © Royal Geographical Society (with The Institute of British Geographers) 2007

2 Boykoff clear view of the state of scientific understanding of climate change and has been corroborated by numerous statements from national science academies and other scientific organisations. Moreover, a 2004 study of peer-reviewed scientific research on climate change found unanimous agreement about the presence of a detectable human ‘signal’ (Oreskes 2004a). While acknowledging that this scientific consensus is not the ‘truth’ translated, this ‘policy-relevant’ information provides a critical input to national and international climate policy. Such solidified discourse on anthropogenic climate change has helped to shape institutional considerations of policy alternatives and their accompanying discursive frames and ‘storylines’ (Hajer 1995). In national contexts, however, divergent climate policy priorities and stances have contributed to complex mosaics of public trust in authority and conflict over decisionmaking (Lorenzoni and Pidgeon 2006). The US federal and UK governments, for example, have both been important actors in international climate negotiations but have played very different roles, the US being branded a foot-dragger, whereas the UK has portrayed itself as a champion of domestic action and international cooperation. Equally, their domestic media have historically taken different approaches towards scientific conclusions on the causes of climate change (Boykoff and Rajan 2007). In combination, the arena of climate policy implementation remains contentious and particularly open to measured analysis of spatial differentiations in news coverage of scientific debates and their influence on national policies (Burgess 2005). When media framing confuses rather than clarifies scientific understanding of anthropogenic climate change, this can create spaces for policy actors to defray responsibility and delay action (Boykoff 2007). Thus, news media coverage plays a significant role in shaping possibilities for future climate policy implementation. In this high-stakes arena of climate science, policy, media and public understanding, there has been a great deal of speculation regarding how this journalistic practice has been used or has ‘disappeared’ from reporting on anthropogenic climate change in recent years. In the following sections, this study interrogates these media practices through content analysis of US and UK newspapers from 2003 to 2006 in order to determine whether ‘balanced’ reporting remains a major contributor to informationally-biased reporting in these key countries, or if we are now flogging a dead norm.

Methods The dataset for the study was composed of newspaper articles from US and UK ‘prestige press’ or ‘quality’ newspapers from 2003 to 2006. The research examined the Los Angeles Times, the New York Times, USA Today, the Wall Street Journal and the Washington Post in the US, and the Independent (and Independent on Sunday), The Times (and The Sunday Times) and the Guardian (and Observer) in the UK. The sample set was accessed and compiled through Lexis Nexis and Proquest/ABI Inform using the key phrases ‘climate change’ and ‘global warming’. In the US, these newspapers are considered as ‘first-tier’ or ‘prestige-press’ news sources, and each has an average daily circulation of nearly one million (Audit Bureau of Circulations 2006). In the UK, these newspapers are also considered to be highly influential, and each has an average daily circulation of over 200 000 (Audit Bureau of Circulations 2007) (see Table 1 for average daily circulation for each newspaper). Through a weighting measure by size of country population, this table provides a measure of the reach and influence of these dailies. While this estimation offers insights into their relative quantitative reach and influence, in terms of qualitative variables (such as type of readership), previous research has also identified these sources as major influences on policy discourse and decisionmaking at national and international levels (McChesney 1999; Doyle 2002), with policy actors routinely monitoring these sources for salient aspects of contemporary public discourse, including climate science. Moreover, beyond directly reaching their readers, these newspapers also influence news coverage in secondary sources, with other reporters, editors and publishers frequently consulting these ‘broadsheets’ for decisional cues on what is ‘newsworthy’ and repurposing their stories in regional and local print outlets. News coverage in these papers therefore provides opportunities to track the dominant news frames associated with anthropogenic climate change (Carvalho and Burgess 2005; Boykoff and Boykoff 2007). In total, 9465 articles on climate change were published in these newspapers between 2003 and 2006, with 2543 articles in US newspapers and 6922 in UK sources. Beginning in January 2003, the sample consisted of a random selection of every sixth article as it appeared chronologically, making a sample of 1607 articles (17% of the population).1 Through quantitative content analysis, codes were


Flogging a dead norm? 3 Table 1

Average daily circulation per issue for selected US and UK newspapers, 2006

Newspaper

Average daily circulation per issue

Average daily circulation per issue per capita ( × 103)

1 231 318 1 683 855 2 528 437 2 058 342 960 684 375 666 233 058 718 221

4.1 5.6 8.4 6.9 3.2 6.3 3.9 12.0

Los Angeles Times New York Times USA Today Wall Street Journal Washington Post Guardian (and Observer) Independent (and Independent on Sunday) The Times (and The Sunday Times)

Note: The US newspapers circulation is from the first three months of 2006 due to data availability (Audit Bureau of Circulations 2006) and UK newspaper circulation is based on information between 27 November and 31 December 2006 (Audit Bureau of Circulations 2007). For the UK newspapers, the Sunday circulation is weighted 1/7 of weekly figures and USA Today does not have a weekend edition. The per capita figures are estimated by US population of approximately 300 million and UK population figures of approximately 60 million residents.

assigned for varying treatments of anthropogenic climate change in each article. The coding was determined not just through frequency assessments of comments or frequencies of words or phrases. Importance was also placed on the labelling of quoted sources, utilisation of terminology, framing of relevant issues and identification of salient elements in each text, as well as tone and relationships between clusters of messages. Multiple stages of piloting were carried out on this content analysis measure to evaluate assessments of the employment of this journalistic norm. Also accounting for spuriousness, these analyses of US and UK sources produced an inter-coder reliability rate of 93.4 per cent, a level that meets established criteria for acceptable inter-coder reliability.2 It is important to note, nevertheless, that such a quantitative approach has clear limits in terms of the detail and texture it can provide for analyses of meaning and discourse. Therefore, such considerations of climate science– media–policy interactions are necessarily complemented by qualitative approaches such as critical discourse analysis, semiotic analysis and interviews (for examples specific to this arena, see Carvalho 2005; Leiserowitz 2006; Boykoff 2007). More broadly, Fairclough (1995) and van Dijk (1988) provide valuable analytical frameworks for further analyses of how power and ideology weave through discourses over time.

Results Figure 1 summarises the quantity of newspaper articles on climate change in the US and UK by

month across the four years and shows a steady increase in coverage leading up to the end of the study period, marked by a more rapid increase in UK newspaper coverage. During this period, coverage quadrupled in UK newspapers and increased approximately two-and-a-half times in the US. While more is not necessarily better, Figure 1 helps to identify key discursive moments in climate science-policy, as captured through media attention.

Peaks in UK coverage of anthropogenic climate change The two largest increases in coverage in the UK took place during June–July 2005 and September– November 2006. June–July 2005 was marked by two particularly prominent moments that garnered heavy newspaper coverage: the Group of Eight (G8) Summit in Gleneagles, Scotland, and increased scrutiny of greenhouse gas emissions from air travel. The G8 summit was strategically preceded by a joint statement from 11 leading international science bodies – including the UK Royal Society and the US National Academy of Sciences – stating that ‘it is likely that most of the warming in recent decades can be attributed to human activities’ (Joint Science Academies Statement 2005, 1). Many news stories linked this joint statement to the G8 meeting. During this same period, media reports outlined European Commission investigations of a tax on aviation fuel, emissions charges and the potential inclusion of aviation in the European Union Emissions Trading Scheme (see Bailey this issue). This also coincided with the UK summer holiday


4 Boykoff

Figure 1

US and UK newspaper coverage of climate change

season, which spurred further discussions and critiques of ‘carbon offsetting’ in media reports. The second increase in coverage in September– November 2006 can be attributed primarily to a series of key interrelated events. Mid-September marked the UK release of the Al Gore film An Inconvenient Truth. This contributed directly to an upsurge of reporting on climate change through personalised coverage of Al Gore, as well as indirectly as a news hook for covering related climate-change issues. Then, in late September, Britain’s Royal Society took the dramatic step of issuing an open letter to Esso, the UK division of ExxonMobil, requesting it to stop funding groups engaged in deliberate disinformation campaigns to undermine scientific consensus on climate change (Adam 2006). Closely following this statement, Richard Branson made his much publicised ‘donation’ of three billion dollars to renewable energy initiatives and biofuel research. This personalised story was widely reported, being both hailed as a philanthropic act and critiqued as the funds were to be invested in Virgin Fuels rather than being donated to another organisation. Further increases during this period were connected to the much anticipated, discussed and criticised ‘Stern Review’, released on 30 October 2006.3 Intense media coverage of the

Stern Review fed into media attention in the Twelfth Conference of the Parties to the United Nations Framework Convention on Climate Change (COP12) meeting in Nairobi that began approximately a week later.4 The events and issues leading up to the conference boosted already heavy media coverage and linked to articles on public sentiment regarding climate policy action, such as the November ‘Stop Climate Chaos’ rally that attracted thousands of people to London’s Trafalgar Square.

Peaks in US coverage of anthropogenic climate change In terms of US coverage, the largest increase coincided with the end of this second period in the UK – November 2006. This was again associated largely with the Stern Review and COP12 in Nairobi, but was further fuelled by connected media coverage of US federal climate policy through the news hook of the mid-term Congressional elections and prominent state-level climate policy action.5 For instance, Arnold Schwarzenegger gained widespread recognition for approving a California bill to cap industrial greenhouse-gas emissions, which helped his re-election campaign (Finnegan 2006). Moreover, when the Democrats took control of the US Senate, Barbara Boxer (Democrat, California) replaced James


Flogging a dead norm? 5

Figure 2

US newspaper coverage of anthropogenic climate change by year, 2003 – 2006, n=421

Inhofe (Republican, Oklahoma) as Chair of the Senate Environment and Public Works Committee. Inhofe had famously declared to the Senate floor (and has repeated many times since) ‘could it be that man-made global warming is the greatest hoax ever perpetrated on the American people? It sure sounds like it’ (Inhofe 2003). In contrast, Boxer has called global warming ‘the greatest challenge of our generation’, and has articulated plans for Congressional legislation to curb anthropogenic greenhouse-gas emissions (Simon 2006, A12). The second largest increase in US coverage in May–June 2006 contributed to climate change becoming a key election issue that November. Chiefly, climate policy rhetoric in the elections was catalysed by heavy media coverage of the May 2006 US release of An Inconvenient Truth. US newspaper reports on the film release spanned several news, business, entertainment and style sections, pushing climate change from an ‘environmental issue’ to one garnering the attention of a wide range of interests and constituents. Such reach was evidenced by a Washington Post ‘Style’ section article covering the documentary premiere (Argetsinger and Roberts 2006) as well as by commentary such as ‘Business World: Warmed Over’ in the Wall Street Journal (Jenkins Jr 2006). During this period the US Supreme Court also agreed to hear the long-awaited case on the Environmental Protection Agency’s (EPA) authority to regulate greenhouse-gas emissions under the Federal Clean Air Act. This case turned on whether carbon dioxide was treated as a ‘pollutant’, and this question – coupled with

increased media attention of Gore’s film – generated an upswing in coverage. Tracking the ebbs and flows of coverage over this timespan provides a foundation for more specific content analysis of media reporting on human contributions to climate change in the US and UK. This quantitative approach produced results that facilitate the identification of ‘critical discourse moments’ where media representational practices may have shifted (Chilton 1987; Carvalho 2005). Carvalho’s discourse analysis of these British ‘quality’ newspapers from 1998 to 2000 defined these moments as those times ‘marked by particular events that potentially challenge existing discursive positions and constructs or, in contrast, may contribute to their further sedimentation’ (2005, 6). Results from these analyses show that the portion of US coverage providing ‘balanced accounts’ of anthropogenic climate change decreased over the period (Figure 2). Statistical tests of difference – using z-scores to compare ratios – were then conducted to determine whether divergences in media coverage from scientific consensus were significant, in other words, whether reporting had perpetrated informational bias regarding scientific consensus through the professional norm of ‘balanced’ reporting. These analyses found that US media representations of anthropogenic climate change diverged significantly from the scientific consensus in 2003 and 2004, but that this was no longer significant in 2005 and 2006 (Table 2). Previous analyses of US newspapers found that coverage from 1990 to 2002 had diverged from the consensus view that humans very


6 Boykoff Table 2 US newspaper discourse and scientific discourse regarding anthropogenic climate change: by year, 2003 – 2006; n=421

Year

Coverage of climate science depicting significant human contribution (%)

‘Balanced’ coverage of anthropogenic climate change (%)

Coverage of climate science depicting negligible human contribution (%)

Was the difference between newspaper coverage and climate science consensus statistically significant?

2003 2004 2005 2006

61.0 89.6 91.8 96.7

36.6 10.4 8.2 3.3

2.4 0 0 0

Yes*** Yes* No No

Note: Newspapers analysed: Los Angeles Times, New York Times, Wall Street Journal and Washington Post. When USA Today was included, the strength of significance did not change. Z-scores per year were: 2003, 7.68; 2004, 2.12; 2005, 1.84; 2006, 1.20, where the numbers represent the percentages of coverage in each year. The significance of divergence of US newspaper coverage from climate-science consensus was determined using z-scores to compare proportions. Z-scores per year were: 2003, 7.73; 2004, 2.22; 2005, 1.92; 2006, 1.31; * p